1. Volume 13, Issue 7, Pages 1055-1067 (July 2005)
X-Ray Crystallographic and NMR Studies of the Third KH Domain of hnRNP K in Complex with Single-Stranded Nucleic Acids 
Paul H. Backe, Ana C. Messias, Raimond B.G. Ravelli, Michael Sattler, Stephen Cusack 
Structure 
Volume 13, Issue 7, Pages (July 2005)
DOI: /j.str
Copyright © 2005 Elsevier Ltd Terms and Conditions

2. Figure 1
Crystal Structures of hnRNP K KH3 Domain Complexed with 15-mer and 6-mer ssDNAs
(A) Ribbon representation of the dimeric assembly of the third KH domain of hnRNP K (red, A; green, B) in complex with the CTC4 (blue in C and D). Some features of the KH domain topology, as well as the hydrogen bonds at the dimer interface, are marked.
(B) The two different DNA dimers formed by crystal contacts seen in the KH3-CTC4 crystal structure. DNA molecules C and D are respectively colored in green and blue. Ribose puckers were as follows: Cyt-1 (Mol C: C3′-endo, Mol D: not visible), Thy-2 (both C2′-endo), Cyt-3 (both C2′-endo), Cyt-4 (both C3′-endo), Cyt-5 (both C3′-endo), Cyt-6 (Mol C: poorly defined, Mol D: C2′-endo).
(C) Ribbon representation of the hnRNP K KH3 domain in complex with the 15-mer ssDNA showing the contents of the asymmetric unit, which consist of three KH3 domains and the DNA molecule. Only two of the KH3 domains (red and yellow) interact with the DNA (blue). The third KH3 domain (green) interacts with the second (yellow) via the β1 strand edges of their respective β sheets (compare with [A]).
(D) Chains A (green) and B (cyan) from the CTC4 complex superimposed onto the uncomplexed KH3 domain (red). The ssDNA (chain C) is colored in blue.
Structure  , DOI: ( /j.str )
Copyright © 2005 Elsevier Ltd Terms and Conditions

3. Figure 2 Detailed View of the KH3-CTC4 Complex
(A) Stereo diagram showing conserved conformation of the central tetranucleotide in the two examples of the KH3-CTC4 complex.
(B) Stereo view showing how Gly400 and Ile403 are inserted like a wedge, preventing the base stacking of Thy2, Cyt3, and Cyt4.
(C) σA-weighted 2Fo − Fc electron density contoured at 1.0 σ (blue) showing the recognition of the first nucleotide, thymine, in the core recognition sequence.
(D) σA-weighted 2Fo − Fc electron density contoured at 1.0 σ (blue) of the interactions of the two middle nucleotides (Cyt3 and Cyt4) in the core recognition sequence with the protein.
(E) Stereo view showing the water-mediated hydrogen bonds and base stacking of Cyt5. Water molecules are indicated as red spheres, and hydrogen bonds are indicated as green dashed lines.
Structure  , DOI: ( /j.str )
Copyright © 2005 Elsevier Ltd Terms and Conditions

4. Figure 3 The KH3 Domain from hnRNP K Binds to the DICE and sDICE RNA
(A) 1H, 15N correlation spectra of KH3 in the absence (black) and presence (red) of sDICE. The backbone amides of Gly406 and Ile423, which shift significantly downfield, are hydrogen bonded to the RNA in the complex.
(B) 1H, 15N correlation spectra of KH3 in the presence of DICE (blue) and sDICE (red). Some of the residues that show larger chemical shift variation between KH3-DICE and KH3-sDICE are highlighted.
Structure  , DOI: ( /j.str )
Copyright © 2005 Elsevier Ltd Terms and Conditions

5. Figure 4
Chemical Shift Perturbations of KH3 upon Binding to DICE and sDICE
(A–C) Average backbone HN variations between the free and the bound KH3. (A) DICE, (B) sDICE, and (C) difference plot ([A] − [B]). The weighted average chemical shift variations were calculated as (([Δ1H]2 + [Δ15N/5]2)/2) for each 1H, 15N pair. Residues 395 and 426 are prolines, and, therefore, no HN backbone can be assigned to them. The backbone resonances of Lys405 and Gly406 in the free KH3 and those of residues Ile423, Asn453, Tyr458, and Gly460 in the KH3-DICE could not be observed.
Structure  , DOI: ( /j.str )
Copyright © 2005 Elsevier Ltd Terms and Conditions

6. Figure 5
Comparison of the Protein-Nucleic Acid Interface in the hnRNP K KH3-ssDNA Crystal and NMR Structures and the Nova-2 KH3-RNA Complex
(A) The NMR solution structure of the third KH domain of hnRNP K interacting with nucleotides Cyt6 and Cyt7 (equivalent to Cyt3 and Cyt4 in the hnRNP K crystal structure).
(B) The third KH domain of hnRNP K interacting with nucleotides Cyt3 and Cyt4 (this work).
(C) The crystal structure of the third KH domain of Nova-2 interacting with nucleotides Cyt13 and Ade14 (equivalent to Cyt3 and Cyt4 in the hnRNP K crystal structure).
Structure  , DOI: ( /j.str )
Copyright © 2005 Elsevier Ltd Terms and Conditions

7. Figure 6
Residues Involved in KH Domain Recognition of Tetranucleotides
(A) Sequences of the core motif recognized by the KH domains of hnRNP K, Nova-2, SF1, and FBP.
(B) Structure-based sequence alignment of KH domains. Residues that are important for specific nucleic acid contacts are highlighted in red. The position of the interacting base in the tetranucleotide sequence is shown on top of the corresponding interacting residue.
Structure  , DOI: ( /j.str )
Copyright © 2005 Elsevier Ltd Terms and Conditions